“The technical tools for handling data have already changed dramatically, but our methods and mindsets have been slower to adapt.”
In 2014, then-United Nations Secretary-General Ban Ki-moon tasked an independent group of expert advisors with providing recommendations on how to achieve a data revolution that would support the UN’s Sustainable Development Goals. Last month, the UN United Nations Statistical Commission and Statistics South Africa co-hosted the first-ever World Data Forum in Cape Town, South Africa.
The forum was based on Secretary-General Ban’s advisory group’s report, entitled A World That Counts, and provided participants from around the globe to share ideas, policies, best practices, and new approaches, and especially to discuss why the acceleration of the data revolution is so critical — especially in developing nations in sub-Saharan Africa.
Are We Delivering?
I didn’t get to attend the World Data Forum, but I was lucky enough to participate in a webinar with a number of data professionals who did attend and were willing to share their thoughts and concerns about topics broached in Cape Town.
Presenters Daniele Doughman and Donatien Beguy (both of the African Population Health Research Center), Rebecca Forman (Center for Global Development), and InKyung Choi (African Centre for Statistics, United Nations Economic Commission for Africa) offered a range of insights focused on the underlying challenges facing the data revolution, and I’d like to discuss some of them here.
Challenges of Data Collection and Use in Africa
A number of presenters touched on the challenges facing those attempting to advance the data revolution in developing parts of Africa, namely, that fundamental data collection needs improvement. Consistent data is necessary on important events and development factors like births, deaths, poverty, trade, taxes, land, environment, education, and so on — factors that you and I blithely assume will be measured by our leaders.
Unfortunately, collection and analysis of this data are hampered by four key factors:
- National statistics offices have limited independence and unstable budgets.
- Misaligned incentives encourage the production of inaccurate data.
- Donor priorities dominate national priorities.
- Access to and usability of data are limited.
So what do we do? And whose job is it to fix this problem?
Governments’ Roles in the Data Revolution
The ability of our national statistics organizations to accurately and regularly collect, clean, and analyze population data free from outside influence is one we in developed countries often take for granted. Where we might even occasionally question the value of national census data to our day-to-day lives, many NSOs in sub-Saharan Africa struggle to collect regular, reliable data.
But why would governments not prioritize statistics collection?
There are a myriad of reasons that leaders in developing countries might not place emphasis on accurate, accessible data collection; some benign, some well-intentioned but misguided and some less honourable.
- In countries facing humanitarian crises, violent conflicts, or environmental disasters, governments may simply not have the resources to dedicate to accurate data collection.
- National statistics offices in some countries lack the manpower, skills, and technology to collect data using methods we are accustomed to — consider that internet access is still non-existent in many areas, and prohibitively slow even when it exists.
- Varying definitions of “open data” lead to disagreement about how accessible NSO information really is — the data may be available to the public, but in a way that makes access difficult or impossible to many.
- Governments may see value in hiding or manipulating data to improve their standing with other world leaders, charity groups, or humanitarian organizations. (Exaggerating stats up or down can alter the level of need or compliance a country may be demonstrating.)
Donors’ Roles in the Data Revolution
One could make the assumption that those funding data collection would prioritize accurate, open data collection and analysis — after all, that very data will be used to measure the impact of their efforts. The problem that lies here, however, is not one restricted to data collection in third-world or developing nations.
The most dedicated, well-intentioned, and fully-funded organizations can make errors when collecting data at the best of times. Those of you participating in the online data journalism course I’m leading with Alberto Cairo right now will recall my example of three different groups approaching a teen suicide crisis, each from their own angle, and all arriving at an entirely wrong conclusion. Unfortunately, without careful collection and review, bias can damage data quality:
- Data can be collected inaccurately — by asking questions that don’t account for cultural differences or biases, or by asking questions of the wrong people.
- Data analysis can rely too heavily on data summaries or not go deep enough, resulting in misleading patterns or inaccurate conclusions.
- Data can be presented in a way that is accidentally unclear — because it lacks context or doesn’t provide enough information
- Data can be presented in a way that is deliberately misleading — to over- or understate the impact of an organization’s work in that area
World Data Forum Solutions: Contracts and Cooperation
One of the webinar presenters spoke about a discussion in a World Data Forum working group that could help address some of these issues: the creation of data contracts between local governments and donor groups that would link funding directly to improving coverage, accuracy, and openness of vital statistics.
This could mean significant improvements to national data collection efforts in these countries: governments would be incentivized to improve NSO operations within their countries and donors would no longer (accidentally or otherwise) influence data collection or analysis. Of course, there are many other ways we can all contribute to the data revolution. Future World Data Forums should focus on:
- Generating policy recommendations for developed countries — how can we provide better resources for NSOs in struggling regions?
- Encouraging participation of statisticians, private sector representatives, government organizations and humanitarian and civil society groups in accelerating progress towards the United Nations’ Sustainable Development Goals
- Investigating ways to use unofficial or citizen-generated data to enrich or supplement official national statistics
- Encouraging data strategy partnerships between countries not normally grouped together because of geography, but who can share knowledge on common situations or challenges
Datassist: Data Science for Nonprofits
At Datassist, we believe the most meaningful change begins with data. But we understand that not every organization has the skills or resources to collect and communicate the data that can tell their story. That’s why we put our years of technical and artistic training at your service — to help you tell the stories that need to be told.
We are proud to work with data journalists and nonprofits of all sizes — right here at home in Toronto, Canada and around the world. Take a look at some of our favourite nonprofit data projects or get in touch with us to discuss how we can help you.